Researchers bet labs won't agree on a single credible third-party safety evaluator
NathanpmYoung · x · 2026-09-13
Responding to @suchenzang's claim that AI labs will place near-zero odds on agreeing to a single third-party safety evaluator, Nathan Young offered a bet. The argument: no body would be competent enough to run real frontier evals (beyond repackaging existing ones), financially independent of existing labs, and have the backbone to speak up when things go wrong.
More from Safety
- Tyler Alterman on the hacking panic: more Y2K than COVID, use 1Password and update software — jd_pressman · 2026-10-31
- Security researcher: vibe coding will force software to consolidate around mature products — jd_pressman · 2026-10-31
- Inference-only chip proposal would bar RL rollouts for models in training — wfithian · 2026-10-31
- Every Used Codex and Agents to Survive the Railway Credential Breach — every · 2026-10-31
- Researchers debate whether more inference compute can substitute for training compute — teortaxesTex · 2026-10-31
- OpenAI blocks weapons details for teens, adds auto-ending chats fixated on violence — rohanpaul_ai · 2026-10-31